Evidence for neurotrophic factors associating with TRPV1 gene expression in the inflamed human esophagus


Address for CorrespondenceChien-Lin Chen MD, PhD, Department of Medicine, Buddhist Tzu Chi General Hospital and Tzu Chi University, 707, Section 3, Chung-Yang Road, Hualien, Taiwan.
Tel: +886 3 8561825; fax: +886 3 8577161;
e-mail: harry.clchen@msa.hinet.netK.R.S. and C.H.Y. contributed equally to this manuscript and are therefore to be considered as co-first authors.


Background  Transient receptor potential vanilloid-1 (TRPV1) receptor has been implicated in the mechanism of acid induced inflammation in gastro-esophageal reflux disease (GERD). It has been demonstrated that the increase in nerve growth factor (NGF) and glial derived neurotrophic factor (GDNF) was associated with the increased expression of TRPV1. We aimed to determine whether expression of TRPV1 was increased in severe inflamed human esophagus, and to test the hypothesis whether the expression of TRPV1 was mediated by neurotrophic factors such as NGF and GDNF.

Methods  We compared biopsies taken from the distal esophagus of 15 patients with erosive GERD, 16 asymptomatic patients (AP), and 10 healthy controls. We assessed the biopsies with reverse transcription polymerase chain reaction (RT-PCR) and real-time quantitative polymerase chain reaction (qPCR) for TRPV1, NGF, and GDNF. Immunohistochemical analysis of TRPV1 protein expression was also determined.

Key Results  Transient receptor potential vanilloid-1 mRNA level and its protein expression were significantly greater in patients with erosive esophagitis than AP (P < 0.001) and healthy controls (P < 0.001). Nerve growth factor and GDNF gene levels in the esophageal mucosa were also significantly increased in patients with erosive esophagitis compared with AP and healthy controls (all P < 0.001). Transient receptor potential vanilloid-1 mRNA correlated well with NGF (r = 0.61, P < 0.001) and GDNF (r = 0.58, P < 0.001).

Conclusions & Inferences  These results support the association of NGF and GDNF in the up-regulation of TRPV1 receptors in patients with erosive esophagitis.


Patients with gastro-esophageal reflux disease (GERD) commonly have heartburn and acid regurgitation with occurrence in 20% of the adult population.1 Reflux esophagitis results from excessive acid exposure of the esophageal mucosa to acidic gastric juice or bile-containing duodenal contents via an incompetent lower esophageal sphincter.2 In patients with GERD, it is conceivable that stimulation of sensory afferents by acid refluxate or inflammatory mediators is considered to be the mechanism underlying symptom generation. It has been shown that acid stimulates sensory afferents in the esophagus by activating proton-gated ion channels including transient receptor potential vanilloid receptor 1 (TRPV1) and acid-sensing ion channels.3–5 Elevated TRPV1 expression has been found in sensory fibers of esophageal mucosa in patients with GERD6 and animal models with reflux-induced esophagitis.7

Afferent fibers of the esophagus are carried both in the vagus and in the spinal nerves, and the majority of them are unmyelinated C-fibers or thinly myelinated Aδ fibers which are mostly polymodal sensory fibers encoding different types of the stimulation.8 It is a general notion that polymodal C-fibers can be further divided into two main groups based on their regulation by either nerve growth factor (NGF) or glial cell derived neurotrophic factor (GDNF).9,10 However, both groups of these fibers can respond to similar types of stimulation and most express the capsaicin receptor TRPV1, which transduces chemical and thermal stimuli.11 A recent study has demonstrated in the esophageal mucosa of rat that the increase in NGF was associated with the increased expression of TRPV1.12 In addition, it has been postulated that the expression of TRPV1 might depend on increased NGF and GDNF13 in patients with rectal urgency.13 Therefore, we aimed to determine whether the expression of TRPV1 in the clinical setting using esophageal epithelial biopsy specimens obtained from endoscopy was increased in severe inflamed human esophagus, and to correlate these findings with neurotrophic factors, such as NGF and GDNF.



The GERD patients with heartburn and/or acid regurgitation of at least 6 months duration, and with erosive esophagitis of at least Los Angeles (LA) classification of esophagitis Grade B14 were enrolled in the study. All subjects did not take proton pump inhibitor prior to this study. The control group (control) was healthy volunteers who enrolled from the community by advertisement and were all college students. All healthy controls were free of esophageal symptoms and had no evidence of acute or chronic illness. The patient control group is asymptomatic patients (AP) who came for upper gastrointestinal (GI) endoscopy for clinical indication other than reflux disease and had no reflux symptoms. In addition, both AP and control groups had no evidence of esophageal inflammation as verified by histology. Esophagitis was identified if there was: (i) basal cell hyperplasia >15% of the total epithelium, (ii) increased papillary length >66% of the squamous epithelium, and (iii) infiltration by polymorphonuclear leukocytes or eosinophils, or both.15 The diagnosis of histological esophagitis required at least two of the above findings. Patients with diabetes taking antiepileptics or anticoagulants were excluded. Pregnancy or nursing women were not eligible for the study. Patients with complicated reflux disease such as Barrett’s esophagus or esophageal strictures were excluded. According to the Montreal classification, GERD is a condition that develops when reflux from the stomach into the esophagus causes symptoms and/or mucosal damage;16 however, patients with LA grade A esophagitis were not enrolled in this study due to the fact that they may represent a heterogeneous group of patients with mild esophageal inflammation who could not be differentiated with patients with non-erosive reflux disease. Patients were also excluded if they were unable to discontinue acid suppressive drugs, or had positive Helicobacter pylori during the examination. The study protocol was approved by Ethics Committee of Tzu Chi Medical Center (Taiwan). The informed written consent was obtained from each subject.


During the endoscopic examination, two biopsies of the distal esophagus were taken. Biopsies were taken using standard biopsy forceps from between esophagitis erosions at a fixed position 3 cm above the squamocolumnar junction in order to achieve sample consistency in all GERD subjects. The extent of mucosal damage was noted and assessed using the LA grading system.14 Erosive GERD was defined by the presence of endoscopically detectable mucosal breaks (erosions or ulcer).

Immunohistochemistry  Fresh biopsy samples were formalin-fixed, paraffin-embedded, and sectioned at 5 μm. After sections were de-paraffinized and rehydrated, antigen retrieval was carried with 10 mmol L−1 citric acid, pH 6 at 95 °C for 20 min. Sections were treated with 3% (v/v) hydrogen peroxide (H2O2) for 10 min at room temperature. The sections were washed three times between all subsequent steps in phosphate buffered saline (PBS), pH 7.3. Non-specific binding was blocked by 5% bovine serum albumin in PBS for 1 h. The sections were incubated with sheep polyclonal anti-TRPV1 antibody (1 : 1000, Abcam, Cambridge, UK) for 18 h at 4 °C, followed by incubation with biotinylated rabbit antisheep antibody (1 : 200; Chemicon, Temecula, CA, USA). Sections were then incubated in standard avidin-biotin horseradish peroxidase reagent (Vector, Burlingame, CA, USA) for 30 min at room temperature. Bound antibody was visualized with 0.05% 3-3′-diaminobenzidine tetrahydrochloride (DAB; Sigma-Aldrich, St. Louis, MO, USA) in 0.05 mol L−1 Tris buffered saline, containing 0.01% H2O2. Negative control was done by omitting primary antibodies.

Image analysis  Whole slide digital images for each section were obtained under an Olympus IX 71 inverted microscope equipped with an Olympus DP70 digital camera and software (Olympus, Tokyo, Japan). Levels of TRPV1 protein were analysed by NIH Image J software (National Institute of Mental Health, Bethesda, MA, USA). Five representative sections of each subject were imaged and the areas and densities of the plaques were measured by the software. Briefly, optical densitometry measures were applied to determine TRPV1 protein expression in each subject, which was calculated as being 3.5 standard deviations above the mean of sections of negative control considered as background. The average pixel values above the set background were used to generate the optical densities.

RNA isolation and reverse transcription  All samples from human subjects were underwent endoscopy. In order to stabilize and protect RNA in fresh specimens, biopsy specimens were stored in RNAlater® Solution (Ambion, Austin, TX, USA) at 4 °C after endoscopy. Total RNA was isolated by MasterPureTM RNA Purification Kit (Epicentre, Madison, WI, USA) according to manufacturer’s instructions. In brief, after homogenization, samples were added with the tissue and cell lysis solution of MasterPureTM RNA Purification Kit (Epicentre) followed by incubating at 65 °C for 15 min and cooling on ice for 5 min, and were centrifuged at 12 000 g at 4 °C for 10 min. The supernatants were added with isopropanol and inverted several times and placed at −80 °C for 10 min. We pelleted the RNA by centrifugation at 12 000 g at 4 °C for 10 min and rinsed twice with 75% ethanol. The RNA was resuspended in DEPC-H2O and quantified by spectrophotometry (A260/280). All RNA samples were reverse transcribed into cDNA at the same time with the ImProm-IITM Reverse Transcription System (Promega, Madison, WI, USA) and oligo(dT)20 and random hexamers primers at 70 °C for 5 min. Master mix contained (per sample): ImProm-II 1 × reaction buffer, 3 mmol L−1 MgCl2, 0.5 mmol L−1 dNTP and ImProm-II reverse transcriptase. The cDNA extension from RNA was done at 50 °C for 60 min according to manufacturer’s instructions. Finally, the cDNA was stored at −20 °C.

Real-time quantitative polymerase chain reaction (qPCR)  The primers used in real-time qPCR were designed using Primer Express Software V2.0 (Applied Biosystems, Foster City, CA, USA) based on sequence information from the National Center for Biotechnology Information database. Homo TRPV1 (NM_080704) forward primer, 5′-GAGTTTCAGGCAGACACTGGAA-3′, reverse primer, 5′-CTATCTCGAGCACTTGCCTCTCT-3′; Homo GDNF (NM_199231) forward primer, 5′-CTTGGGTCTGGGCTATGAAAC-3′, reverse primer, 5′-CAAAGGCGATGGGTCTGC-3′; Homo NGF (NM_002506) forward primer, 5′-AGCAAGCGGTCATCATCC-3′, reverse primer, 5′-GTGGCGGTGGTCTTATCC-3′. Real-time qPCR was performed in triplicate by using an ABI PRISM® 7300 Real-time PCR System (Applied Biosystems) with the iTaqTM SYBR® Green Supermix with ROX (Bio-Rad, Hercules, CA, USA) according to manufacturers’ instructions and protocols. Thermal cycling was performed as follows: initial denaturation at 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Homo β-actin was used as a reference; i.e. each sample was normalized on the basis of its β-actin content. The relative change in all target genes expression was determined by the fold-change analysis.

Statistical analysis

All results were expressed as mean ± SEM. Analysis of variance (anova) with a post hoc correction was used to compare differences in the expressions of TRPV1, NGF, GDNF mRNAs and TRPV1 protein level among different groups of the patients and healthy controls. Correlation was tested for significance with a Pearson’s correlation coefficient. A P-value of less than 0.05 is defined as significant.


Between January and August 2008, fifteen patients with erosive esophagitis (six women, mean age 52.7 years, range 34–78 years) (LA Grade B in eight, Grade C in six, Grade D in one) and 16 asymptomatic patients (eight women, mean 42.8 years, range 18–58 years) entered the study. Ten healthy controls (five women, mean age 24, range 22–30 years) were also included in this study. No age difference was found between the patient groups, but there was a significant age difference when healthy controls were compared with AP and GERD groups (both < 0.05). Increased papillary length was present in 12 of 15 GERD patients (LA Grade B in five, Grade C in six, Grade D in one), whereas basal cell hyperplasia was observed in 12 of 15 GERD patients (LA Grade B in five, Grade C in six, Grade D in one). Mucosal infiltration by polymorphonuclear leukocytes was present in 11 of 15 patients (LA Grade B in seven, Grade C in three, Grade D in one).

The TRPV1-immunostained sections of esophageal biopsies taken from AP, controls and GERD were shown in Fig. 1. The sections from GERD group were divided into two groups without (Fig. 1A) and with (Fig. 1D) incubation by TRPV1 antibody, i.e. negative control. The density for TRPV1 protein was stronger in GERD patients, (Fig. 1D) lighter in healthy control (Fig. 1B) and AP (Fig. 1C), and lightest in negative control (Fig. 1A). The optical densities of the TRPV1 protein levels was significantly greater in GERD patients than AP (146.50 ± 2.88 VS 86.47 ± 2.99, < 0.001) and healthy controls (146.50 ± 2.88 VS 89.74 ± 2.71, < 0.001) (Fig. 2). However, optical densities of the TRPV1 protein levels did not differ between AP and healthy controls (86.47 ± 2.99 VS 89.74 ± 2.71, > 0.05) (Fig. 2).

Figure 1.

 Representative images of TRPV1 immunostaining of histological sections in (A) a negative control (NC), (B) a healthy control (Controls), (C) an asymptomatic patients (AP), and (D) a GERD patient.

Figure 2.

 Optical densities of TRPV1 protein expression were significantly higher in patients with GERD compared with asymptomatic patients (AP) (*< 0.001) and healthy controls (*< 0.001). Bar represents the mean value ± SEM.

The TRPV1 mRNA expression was significantly greater in patients with erosive esophagitis than AP (2.02 ± 0.23 VS 1.09 ± 0.06, < 0.001) and healthy controls (2.02 ± 0.23 VS 1.00 ± 0.05, < 0.001) (Fig. 3). The TRPV1 mRNA expression increased with the severity of GERD [1.6 (0.98–2.7), LA grade B; 2.3 (1.21–3.59)], although there was only one patient with LA grade D whose TRPV1 RNA level was 3.67. Expression of NGF and GDNF gene level in the esophageal mucosa was markedly greater in patients with erosive esophagitis compared with control and AP groups (all < 0.001) (Figs 4 and 5). The expression of TRPV1 gene correlated well with NGF (= 0.64, < 0.001) (Fig. 6A) and GDNF (= 0.63, < 0.001) in esophageal mucosa (Fig. 6B).

Figure 3.

 Analytical data of TRPV1 expression in the esophagus. TRPV1 mRNA expression was significantly higher in patients with GERD compared with asymptomatic patients (AP) (*< 0.001) and healthy controls (*< 0.001). Line represents the mean value.

Figure 4.

 Analytical data of NGF expression in the esophagus. NGF mRNA expression was markedly greater in patients with GERD compared with AP (*< 0.001) and healthy controls (*< 0.001). Line represents the mean value.

Figure 5.

 Analytical data of GDNF expression in the esophagus. GDNF mRNA expression was significantly higher in patients with GERD compared with AP (*< 0.001) and healthy controls (*< 0.001). Line represents the mean value.

Figure 6.

 Correlation between mRNA expressions of TRPV1 and neurotrophic factors, such as NGF and GDNF in all subjects. The expression of TRPV1 RNA correlates well with NGF (= 0.64, < 0.001) (A) and GDNF (= 0.63, < 0.001) in esophageal mucosa (B). Each data point represents an individual subject. Black circle, white circle, and inverted triangle indicate healthy subjects (Controls), AP and GERD patients, respectively.


Our study reconfirmed the previous investigation by demonstrating in human esophageal mucosa that patients with erosive esophagitis exhibited markedly greater TRPV1 protein and its gene expression than AP or control group.6 Patients with erosive esophagitis were also characterized by significantly greater gene expressions of NGF and GDNF in the esophageal mucosa when compared with AP and control groups. In addition, we have shown increased expression of TRPV1 gene correlated well with NGF and GDNF mRNA expression in esophageal mucosa. Therefore, our study would suggest that acid induced inflammation in the esophagus may up-regulate TRPV1 receptors, which is associated with increased expression of NGF and GDNF.

In a previous investigation on patients with rectal hypersensitivity and fecal urgency,13 Chan et al. have demonstrated that increased TRPV1 expression was associated with increased NGF and GDNF expressions in the mucosa layers of the patient group, suggesting that neurotrophic factors may act in proliferation and expression of visceral afferents and therefore mediate the up-regulation of TRPV1 receptors. Similar findings have been reported in animal models17 and patients with irritable bowel syndrome.18 Together with our findings, it is conceivable that the mechanism responsible for increased TRPV1 expression in patients with GERD are likely to be mediated by the release of NGF and GDNF, which can be produced locally during repeated esophageal inflammation. However, other study did not reveal any difference in peripherin expression of the inflamed esophagus between reflux patients and healthy controls, indicating increased TRPV1 receptor expression without significant increase of nerve fibers.6 The cause of these discrepancies are unclear, but may be partially attributed to different methods and patients studied. Moreover, it has been demonstrated that inflammatory mediators may sensitize TRPV1 receptors.19 Likewise, TRPV1 expression increases during inflammatory processes associated with pain.20 It is conceivable that the presence of esophageal inflammation in patients with erosive esophagitis may also play a role in TRPV1 expression within the mucosa as observed in our study.

The mechanisms through which NGF and GDNF enhances TRPV1 expression remain unclear. It has recently been shown that both neurotrophic factors can activate the expression of the small GTPase protein Ras, which is important for modulating TRPV1 expression.21 In addition, NGF or GDNF is likely to regulate TRPV1 expression by influencing membrane potential, as both are shown to result in a depolarizing shift in membrane potential via the TTX-resistant sodium channels.22,23 We acknowledged the importance of further analysis of NGF and GDNF behind their gene levels, i.e. protein expressions, which would help further elucidate the mechanisms involved in the current findings. However, our finding of greater correlation between TRPV1 and neurotrophic factors may be more relevant to the association but not the causation as observed in other studies,13,17,18 Therefore, such notion requires further studies for confirmation.

The mucosa of the esophagus is innervated dually by vagus and spinal afferents which are exquisitely sensitive to light touch, pH and chemicals.24 These afferents are suggested to play the thermo- or chemosensory functions.25 Acid-induced mucosa injury and inflammation can alter the structure of primary sensory neuron in the esophagus of patients with erosive GERD.6,26 This notion has been further supported by a recent study in non-erosive GERD, which suggested that chronic esophageal inflammation without macroscopic injury may also influence the structure and function of mucosa afferents by the release of local inflammatory mediators.27 Furthermore, the presence of TRPV1 immunoreactive fibers in the esophagus is believed to play a role in mediated nociception.28

Recent studies implied that the NGF and GDNF are important on regulation of GI function.29–31 It has been demonstrated that chronic stress induces visceral hypersensitivity by increasing the expression of NGF in the colon wall of rats.31 In addition, GDNF is essential for the development of the enteric nervous system from GDNF knockout studies.32 A reduction in GDNF protein has been observed in the mucosa of human aganglionic bowel in Hirschsprung’s disease.29 GDNF is reported to be up-regulated in human inflammatory bowel disease and has strong antiapoptotic properties in colonic epithelial cells of rats.30 Although the mechanism of these neurotrophic factors in human esophageal mucosa is yet unclear, these studies may suggest the importance of NGF and GDNF in regulating normal esophageal functioning.

The probable origins of the increased expression of TRPV1 receptors are likely to be intrinsic and extrinsic to the esophagus. Results of recent studies by using immunohistochemical studies in human esophagus have indicated that the fibers positive for TRPV1 represented intrinsic afferent neurons involved in local reflexes.6 In this study, the presence of increased gene expression of TRPV1 along with its protein expression in the esophageal mucosa of patients with erosive esophagitis may provide an additional evidence for an intrinsic origin, as it has been reported that esophagitis causes an over expression of TRPV1 in human esophagus6,26 and rat vagal and spinal neurons.6,26 The extrinsic evidence for such nerve fibers is supported by the existence of TRPV1 in a subset of dorsal root ganglion cell bodies and nerve fibers in a rat model of reflux esophagitis.7 In addition, similar findings have been demonstrated in feline models by experimental esophageal acidification.33

There are potential mechanisms by which TRPV1 may contribute to the development of esophagitis. Activation of TRPV1 receptors by acid can release neuropeptides such as SP and calcitonin gene-related peptide34 from the nerve terminals of primary sensory afferents to play an important role in plasma extravasations and inflammation. In addition, inflammation itself can further induce the release of several inflammatory substances including the arachidonic acid derivative anandamide, lipoxygenase derivatives, and leukotriene etc. These substances have been shown to be potent agonists of TRPV1 receptors.35,36 Therefore, activation of TRPV1 receptors by these inflammatory substances acts as a positive feedback loop to enhance the release of neuropeptides from the nerve terminals and other pro-inflammatory cytokines.13

Upregulation of TRPV1 genetic expression in esophageal mucosa of patients with erosive esophagitis, suggests that drugs that antagonize endogenous inflammatory substances that activate this receptor could potentially lead to new therapies for heartburn perception and reflux-related esophageal inflammation. However, conflicting data concerning agonist and antagonist treatment, both being protective, were reported under different experimental conditions. It has been previously reported that a novel TRPV1 channel blocker, JNJ 10185734, attenuates the severity of dextran sulfate sodium induced colitis in mice, implying that TRPV1 may contribute to the initiation or maintenance of inflammatory process in the gut.37 On the other hand, applying the dinitrobenzene sulfonic acid model of intestinal inflammation in mice, the protective effects of curcumin were investigated and the TRPV1 receptor was identified as playing a major role in the observed protection.38 In addition, mice lacking the TRPV1 receptor were found to exhibit a much more severe colitis, supporting the pharmacological concepts of TRPV1 receptors being crucial regarding protection against inflammatory situations in the colon.39

The esophageal biopsies were taken according to the method by Newton et al.26 It would be of additional interest to evaluate the degree of inflammation histologically, although it has been recently demonstrated that the histological grading of esophagitis correlated well with endoscopic findings. Therefore, the clinical grading applied in our study would be ideally adequate to show that our patients had significant GERD.40 As esophageal epithelial changes in response to acid reflux are patchy, specimens were always taken a constant distance proximal to the squamocolumnar junction and areas of visible erosive lesions were avoided to minimize the sample error. The other potential limitation of this study is the small sample size which could influence the strength of our findings, although our findings support the notion that the increased expression of TRPV1 is associated with the severity of esophagitis based on LA grading. However, the previous study which demonstrated the increase of TRPV1 in human esophageal mucosa in patients with erosive esophagitis also did not avoid the overlap in TRPV1 expression between control and GERD patients,6 and so did our study. In our study, healthy controls were enrolled from subjects younger than asymptomatic patients and GERD patients. The possibility of these target genes influenced by age can not be excluded, because the elders had higher levels of gene expression. Larger studies with age-control and adequate sample size will help address current findings.

In summary, we have found evidence for an increase in TRPV1 mRNA and protein expressions in the mucosa of biopsy specimens from the inflamed esophagus in patients with erosive esophagitis. The present study supports a potential mechanism in which NGF and GDNF might accompany the upregulation of TRPV1 receptors leading to inflammatory situations in the esophagus of patients with GERD. The development of pharmacological agent targeting the TRPV1 channel for GERD-induced esophagitis has yet to be investigated.


This study was supported by a grant, TCRD-9721, from Buddhist Tzu-Chi General Hospital, Taiwan. All investigators contributed to the design, interpretation of data and to the writing of the manuscript. C.L. Chen was mainly responsible for the writing of the paper and study design. K.R. Shieh was the principal investigator and responsible for genetic methodology and data analysis regarding TRPV1 proteins. T.T. Liu and C.H. Yi were responsible for endoscopy and data collection. H. L. Tseng and H.T. Hsieh were responsible for data analysis regarding genetic results. H. C. Ho was responsible for tissue sectioning and immunohistochemical staining.